Tritium ($^3H$ or T) has been produced mostly by atmospheric nuclear weapon tests, and entered the ocean in the form of water (HTO). As tritium exists as water itself, it has been regarded as an ideal tool to study the transport of water masses. In April 2001 we collected water samples in the western Japan Basin (WJB) for tritium and helium measurement. The timely sampling provided direct evidence of the bottom water formation, resulting in the drastic increase in tritium concentration from 0.3 TU in 2000 to 0.67 TU in 2001. Considering that the new bottom waters were found mostly in the WJB, it implies that maximum 1% of the whole bottom layer below 2600 m should be replaced with the surface water during the severely cold winter 2000 2001. $^3H-^3He$ age, showing the elapsed time since the water left from the surface, can be used to calculate oxygen utilization rate by dividing AOU by the age. Under the condition of 90% oxygen saturation in the surface water, the integration of OUR in the water column below 200 m yields net oxygen consumption of 12 mol $(O_2)m^{-2}yr^{-1}$, which corresponds to the export production of $99;g;C;m^{-2}yr^{-1}$. This estimate is comparable to a previous estimate based on satellite data and implies that the ratio of export to primary production(f-ratio) is as high as 0.5 in the WJB.